Modelling and analysis of user behaviour

Abstract

This paper provides a thorough study of modeling and analyzing user behavior in online shopping. The research is based on simulated interactions involving 40 users with an e-commerce platform while selecting different gaming monitors. Key behavioral indicators such as interaction time, specification views, and filter usage were standardized to allow for accurate clustering. Using hierarchical cluster analysis with Ward’s method and Euclidean distance, three distinct user groups were identified, each demonstrating unique behavioral patterns toward BenQ, Asus, and Xiaomi products. Discriminant analysis was utilized to evaluate the classification's effectiveness and to identify the most significant variables affecting group differentiation. The findings showed an accuracy, confirming the reliability of the clustering method. This research emphasizes the significance of analyzing behavioral data to tailor user experiences, improve marketing tactics, and boost digital product usability. The findings offer important insights into how data-driven decision-making can enhance adaptive e-commerce solutions.